With the worldwide web at your fingertips, it can be very easy to become overwhelmed with all of the dog training resources available, and the different training methods that exist. It makes it especially confusing when the different methods contradict each other, with well-known people in the field making it seem like their method is the very best. However, when it comes down to it, there are two main methods of training dogs, of which I will highlight the basics.
Aversive Training: discouraging undesired behaviors with the use of an aversive/unpleasant stimulus; encouraging desired behaviors by taking the stimulus away. These include but are not limited to: leash corrections and leash pops, shock collars, ear pinches, prong collars, and alpha rollovers.
Positive Reinforcement-Based Training: adding something that a dog likes following a behavior in order to increase the likelihood that the behavior will occur again. These include: treats, play sessions with toys, and secondary reinforcers such as praise or a clicker.
Now, you will surely encounter people and trainers that say that aversive training methods work if done correctly, and that dogs learn more quickly when learning this way than with any other method. However, what can often happen is that dogs are responding out of fear, and they still do not know what is expected of them. This can result in a more generalized anxiety (say, a dog who doesn’t know why he is receiving shocks from a shock collar), or possibly in more negative behaviors (for instance, if a dog is given a shock when chewing on something inappropriate like a shoe, they still do not know what is appropriate to chew on, and may move on to another household item). There is a common saying in dog training: rather telling what a dog what they can’t do, show them what they can. This is where positive reinforcement comes into play: is your dog chewing on the table leg again? Well, you could give them a shock from a shock collar, or yell, among many other aversive methods. But this still does not give your dog the necessary information about what they should be doing. Instead, you can employ positive reinforcement by interrupting the dog and redirecting them to chewing on something appropriate and enjoyable, which will reinforce chewing on dog-friendly items like toys.
Aversive methods can also be risky. There is a method called an “alpha rollover”, which consists of forcefully pushing a dog onto their back so they are laying in a submissive position, the (incorrect) theory being that the dogs will then be submissive and see the person as their “pack leader”. However, dogs are more likely to feel very afraid (which is not the same thing as submissive), and potentially defend themselves by snapping or even biting. When they do defend themselves, one of two things can happen: 1. the person backs off and gives the dog their space, which reinforces the biting or snapping and makes it more likely that the behavior will occur again (“hey, it worked!”), or 2. The person then punishes the dog for defending him or herself, teaching the dog that their attempts to communicate are not being heard, and to be even more afraid.
But enough of the negativity: let’s discuss what is so great about positive reinforcement. First of all, it’s enjoyable to the dogs! You can see the wheels turning when a dog is learning a new trick: “oh, when I do this, I get a treat? Wow! I can just keep doing it and I get more yummy treats? This is great!” Second of all, it’s fun! It’s obvious that many dogs, when training using positive reinforcement methods, really like learning. My dog Sasha is a great example of this: she’s a smart girl, and when I am teaching her a new trick, she remains hyper-focused on me while trying to figure out what is expected of her. I can tell once she has a breakthrough, because she will start happily offering the behavior to me in hopes of more treats.
This leads me to a question I get a lot: Isn’t it bad that we are giving so many treats? Are we spoiling our dogs?
Well, the beauty of positive reinforcement training is that, after the initial learning period, you don’t have to give treats as often. I like to compare it to the slot machine/vending machine analogy:
- When teaching a new trick, we want to be like a vending machine, and predictably give treats each time we see a desired behavior so that the dog learns exactly what we want from them. This helps to polish the behavior and have the dog reliably offer it every time.
- Once the dog has successfully mastered the command/behavior, we want to be more like a slot machine: in other words, unpredictable. If we start offering treats in a pattern of intermittent reinforcement (say, only give treats for every 2 out of 5 sits that a dog offers), this strengthens the behavior because they never know when they will be getting a treat, and will keep trying even when they don’t get a treat each time. Compare this to a vending machine, where if it doesn’t dispense your candy bar as expected, you get frustrated.
Additionally, we don’t always have to use treats as a reward. Some dogs are very motivated by a quick game of tug with a rope toy, for example, or a brief chew on a squeaky toy. In training classes, we teach the dogs that when they come when they are called, they get to chase us for a few paces (which is a reward in itself because it’s fun!).
I could go on and on, but when it comes down to it, this is the question that I pose when people come to me wondering which method to use: what kind of relationship do you want to build with your dog? Do you want to build a relationship that is built on “having power over” your dog, in which the dog likely responds out of fear of something unpleasant happening if they don’t listen to you? Or would you rather use a method that is fun for your dog, which makes them excited to listen and eager to learn? The answer to that seems clear to me.